|Publication number||US7894651 B2|
|Application number||US 11/681,345|
|Publication date||Feb 22, 2011|
|Filing date||Mar 2, 2007|
|Priority date||Mar 2, 2007|
|Also published as||CA2679837A1, EP2115664A1, EP2115664A4, US20080214907, WO2008109421A1|
|Publication number||11681345, 681345, US 7894651 B2, US 7894651B2, US-B2-7894651, US7894651 B2, US7894651B2|
|Inventors||Dina Gutkowicz-Krusin, Joseph V. Gulfo|
|Original Assignee||Mela Sciences, Inc.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (12), Non-Patent Citations (31), Referenced by (7), Classifications (10), Legal Events (3)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This disclosure relates to quantitative analysis of skin characteristics.
Skin lesions, for example, include tissue on or in the skin that has abnormal characteristics. In general, skin lesions are malignant, pre-malignant or benign and also can be categorized based on physical characteristics. Some skin lesions are pigmented and others are non-pigmented. Examples of pigmented malignant and pre-malignant lesions include melanoma, some basal cell and squamous cell carcinomas and actinic keratoses. Examples of pigmented benign lesions include low-grade dysplastic nevi, congenital nevi and seborrheic keratoses.
Dermatologists or other physicians determine whether a particular lesion on a patient's skin should be biopsied to rule out melanoma. The determination is especially difficult for early stage melanoma lesions which are similar to many benign pigmented lesions. Analysis by a dermatologist of malignant and benign lesions entails subjective visual assessment of lesion characteristics.
In one aspect, an apparatus for analyzing skin characteristics includes a device to acquire a digital image of an area of skin and a processor to quantitatively analyze the digital image. The processor determines a characteristic of all or part of the area of skin that is indicative of a skin condition of interest and, depending on results of the quantitative analysis, provides information to a user about the area of skin relative to the condition of interest. The information includes an indication to the user as to whether a knowledgeable person should evaluate the area of skin.
Implementations may include one or more of the following features. For example, the information may include a classification of the skin condition of interest. The information may not be necessarily sufficient for a definitive medical diagnosis.
The quantitative analysis may include comparing the characteristic of all or part of the area of skin to a threshold associated with the condition of interest.
The characteristic indicative of a skin condition of interest may have a high specificity and high sensitivity in a general population. Both the specificity and sensitivity may be greater than 90 percent in the general population. The characteristic that is indicative of a skin condition of interest may not be recognizable by visual inspection by a physician.
Part of the area of skin may include a pigmented skin lesion. The condition of interest may include malignancy. The quantitative analysis may suggest that malignancy can or cannot be ruled out.
The processor may provide an indication to a user that it is not necessary to have the area of skin evaluated by a knowledgeable person. The processor may provide an indication that it is advisable to have the area of skin evaluated by a knowledgeable person. The user may not be a licensed health care provider or a physician. The knowledgeable person may be a physician and the user may be someone other than a physician. The knowledgeable person may be a dermatologist and the user may be a physician other than a dermatologist.
The apparatus may include an audible or visible indicator that provides the information about the area of skin relative to the condition of interest.
The device may acquire, at different times, digital images of the area of skin and quantitatively analyze the different digital images. The processor may compare the results of the quantitative analysis for each image and, based on the comparison, indicate whether a change in a condition of the area of skin has occurred.
The device may be configured to be applied to the skin. The device may be triggered by the user to acquire the digital image. At least part of the apparatus may be handheld.
In another aspect, a quantitative analysis is performed on an image of an area of skin to determine a characteristic of all or part of the area of skin that is indicative of a condition of interest. Information may be provided to a user about the area of skin relative to the condition of interest, based on the quantitative analysis. The information includes an indication to the user whether a knowledgeable person should evaluate the area of skin.
The information may not be necessarily sufficient for a definitive medical diagnosis.
Images of the area of skin may be acquired at different times and a quantitative analysis on each of the images may be performed to determine, for each image, a characteristic of all or part of the area of skin. The results of the quantitative analysis may be compared for each image. Information may be provided to the user, based on the comparison, that indicates whether a change in condition of the area of skin has occurred.
The knowledgeable person may be a physician and the user may be someone other than a physician. The knowledgeable person may be dermatologist and the user may be a physician other than a dermatologist.
In another aspect, a person's skin is quantitatively determined whether it is characterized by a condition of interest, based on an image of the person's skin and statistical information about skin of people in a general population. The quality of the determination is sufficient to reliably indicate whether examination by an expert of the skin relative to the condition of interest is desirable.
In another aspect, an image of a region of skin of a person is automatically analyzed and an indication that a condition of the region of skin should be analyzed further by a human expert is automatically provided.
In another aspect, qABCD parameter values of images of skin lesions are quantitatively determined. Based on statistical information from a general population, lesions that are non-melanoma and lesions that cannot be ruled out as melanoma are determined with a high sensitivity and specificity. An indication of the result of the determination, which is based on statistical information, is provided.
In another aspect, qRING parameter values of images of skin lesions are quantitatively determined. Based on statistical information from a general population of humans, lesions that are non-melanoma and lesions that cannot be ruled out as melanoma are determined with a high sensitivity and specificity. An indication of the result of the determination, which is based on statistical information, is provided.
Implementations may include one or more of the following features. Each qABCD parameter value may be compared to a corresponding threshold value. A qABCD characteristic may be determined as present if the corresponding qABCD parameter value is greater than the corresponding threshold value. A qABCD characteristic may be determined as absent if each qABCD parameter value is less than the corresponding threshold value.
Each qRING parameter value may be compared to a corresponding threshold value. A qRING characteristic may be determined as present if the corresponding qRING parameter value is greater than the corresponding threshold value. A qRING characteristic may be determined as absent if each qRING parameter value is less than the corresponding threshold value.
Each threshold value may be derived from a reference database of imaged skin lesions in which each threshold value maximizes a diagnostic accuracy of a corresponding qABCD or qRING parameter value. An indication of the advisability of having the skin lesions evaluated by a knowledgeable person may be provided if one or more qABCD characteristics or qRING characteristics is present. The sensitivity and specificity of the qABCD characteristics and qRING characteristics may be greater than 90 percent.
In another aspect, an apparatus includes a camera for acquiring a digital image of a region of skin that has a skin lesion. A processor connected to the camera is programmed to quantitatively analyze the digital image to determine the presence of a characteristic, indicative of malignancy, in the skin lesion. The processor may provide an indication to a user to have the skin lesion evaluated by a knowledgeable person if the characteristic is present. The processor may provide an indication to the user that it is not necessary to have the skin lesion evaluated by a knowledgeable person if the characteristic is not present. The apparatus also includes a display for viewing the skin lesion image and indication provided by the processor.
Other features and advantages will be apparent from the description and from the claims.
An example of a way to analyze skin lesions or other skin conditions using a skin characteristics analysis device 2 (which, for this example, we sometimes call a lesion analysis device) is shown schematically in
Traditionally, a person interested in knowing whether a skin lesion on his or her body is malignant will visit a medical professional who proceeds to examine the lesion. The medical professional will decide whether this lesion is in need of a biopsy to rule out malignancy. According to the present disclosure, however, that person may first obtain information which indicates whether an evaluation by the medical professional is necessary.
As shown in
In some implementations, the device 2 can be used by a primary care physician or other physician that does not specialize in dermatology. For example,
The user 4 applies the lesion analysis device 2 to a region of skin that includes the skin lesion and activates the device (101) to cause a digital image of the lesion to be obtained. The imaged lesion may either be pigmented or non-pigmented. The lesion analysis device 2 then performs a quantitative analysis (103) of the digital image to determine the presence or absence of one or more characteristics indicative of the lesion condition. As a result of the analysis, the device 2 provides information to the user (105) relating to the lesion condition. For example, the device 2 may notify the user 4 that the lesion is unlikely to be malignant or that the user should have the lesion further evaluated by a medical professional, such as a dermatologist. The information also can include a reminder that lesions with changing features should be seen by a medical professional regardless of the results provided by the device 2.
An example of the lesion analysis device 2 is shown in
All the hardware components to the device, including the camera 10, attachment 12, and computer 20 can be off-the-shelf components, i.e., they are ready-made for a variety of uses and available for sale, lease, or license to the general public. Accordingly, an inexpensive lesion analysis device may be produced due to a reduction in assembly time and costs. Other implementations for acquiring and storing the images may be used as well. The lesion analysis device 2 may have access to other computers 24 or servers 26 through a network 22, such as the Internet. The device 2 may be connected to the network 22 by means of a variety of network connections such as a phone line, a cable or a wireless link.
The light travels through the glass plate 18, is reflected off the skin, travels back (203) through glass plate 18 and then passes (205) the polarizer 16 so that specular reflection is minimized. The polarized light then is acquired by the digital camera 10, which produces a digital image of the region of skin in at least three channels (R, G B). The digital image can be saved on the camera 10 or on the computer 20 for storage, viewing and processing. In either case, the digital image, or a copy of it, is passed from the camera 10 to the computer 20.
The computer 20 has a digital input 20 a for receiving the digital images from the camera 10. The digital input 20 a can include any conventional device that allows data to be transferred between the camera 10 and computer 20. For example, the digital input 20 a may be a universal serial bus connection. Alternatively, the digital input 20 a may be a wireless connection. Other types of digital inputs may be used as well. The memory 20 b serves to store the digital images received from the camera 10 as well as software that is used to control the image analysis. Examples of memory 20 b include the computer's hard drive, read only memory, random access memory, floppy disks, CD-ROM, or DVD-ROM. Other suitable media may be used as well.
The processor 20 c, under the control of the software, performs image analysis on the received image and provides the results of the analysis to display 20 d. Examples of the display 20 d include cathode ray tube monitors, liquid crystal display monitors and touch-sensitive screens. Other displays may be used as well. Example screen shots of what a user would see on the display 20 d are shown in
Additional lesion information, which does not relate to screening for malignancy, may also be provided. For example, the device 2 may indicate the sub-classification to which a benign lesion likely belongs (such as seborrheic keratosis, melanocytic nevus, etc.). Furthermore, if a user is unsure whether a lesion is increasing in size or changing color, the device 2 may be used to store images at successive points in time to ascertain lesion characteristics over a period of time. The results of quantitative analysis for the different images may then be compared. If the analysis indicates that a lesion has changed, the device 2 notifies the user 4 and suggests further evaluation by a physician (see
In some implementations, the device 2 also may enable payments based on each instance the device is operated. For example, a user may enter a patient's account information into the device and, after a predetermined number of device operations (which include image acquisition and analysis), the account information may be transferred over the network to the server 26 where the charge for use is processed. Other ways of paying for utilizing the device 2 may be incorporated as well.
The steps carried out by the processor 20 c to determine whether the region of skin contains a benign lesion or one that is in need of further evaluation can include (1) image calibration, (2) image quality control, (3) lesion segmentation and (4) lesion analysis, as discussed in the next section.
Flat-fielding also establishes a proper color balance for the system. The white calibration image may be obtained by imaging a white target of uniform reflectance in order to have a measure of the non-uniformities introduced by the device. The calibration image can be stored in the memory of the computer prior to first use so that the user is not required to obtain calibration images. The white calibration target may also be provided to the user for obtaining calibration images.
Once the received image has undergone calibration, the image is analyzed for quality (405) in an image quality control step. For example, the received image is checked for overexposure, underexposure, whether a lesion is present, whether the lesion is fully in the field-of-view of the camera, whether one or more dark spots are due to normal skin features or hair, and whether those spots can be easily removed through further image processing. Other checks for image quality may be implemented as well. If the processor 20 c determines that the image fails the quality checks, the processor may display (406) a report to the screen indicating that the image needs to be re-taken. Alternatively, the processor may display a report that the image cannot be analyzed.
Following image quality analysis, the image is segmented (407) using segmentation algorithms. Examples of segmentation algorithms are described in U.S. Pat. No. 6,208,749, incorporated here in its entirety. In general, image segmentation entails separating the received image into regions that belong to a lesion and regions that are normal skin.
During lesion analysis, the software may instruct the processor 20 c to analyze the lesion image based on a set of parameters corresponding to clinical characteristics. After performing the analysis, the processor may provide information to the user that is indicative of the lesion condition. For example, the processor 20 c may determine that the lesion is benign or that there is insufficient data for a definitive medical diagnosis and subsequently post (413) the results of its calculation on the display 20 d.
Examples of pigmented lesion characteristics useful in evaluating a lesion's condition include a set of four parameters known as ABCD characteristics. Traditionally, these four characteristics are determined visually and provide a subjective means for physicians and patients to identify pigmented skin lesions that could be melanoma. The four parameters represented by the ABCD characteristics are lesion asymmetry (A), border irregularity (B), color variegation (C) and lesion diameter (D).
Although the ABCD characteristics are typically assessed visually, it is also possible to define and quantify those characteristics so as to provide an objective means of analyzing pigmented skin lesions for melanoma. Quantitative ABCD characteristics (called qABCD) corresponding to the ABCD characteristics may be obtained by applying a set of mathematical algorithms to each segmented digital image that contains a lesion. In the general population, having at least one qABCD characteristic has a high sensitivity to melanoma and, at the same time, a high specificity, in which sensitivity is the probability that melanoma is classified correctly in a patient and specificity is the probability that a benign lesion is correctly classified in the patient.
Among lesions biopsied by a dermatologist, however, qABCD characteristics may not provide complete discrimination of melanomas from other pigmented skin lesions. Therefore, qABCD characteristics, alone, may not be sufficient for a medical diagnosis of melanoma, but remain useful as a first identification of pigmented skin lesions which require further evaluation.
Accordingly, in some implementations, the device 2 may be programmed to determine whether qABCD characteristics are present in an imaged pigmented lesion. The processor 20 c computes (409) a value (called a qABCD value) for each qABCD characteristic from a corresponding mathematical algorithm. Each qABCD value is then compared (411) to a corresponding threshold value. If the qABCD value is above the threshold, a qABCD characteristic is determined to be present. If the qABCD value is below the threshold, the qABCD characteristic is determined to be absent. Should the processor 20 c establish that a minimum of one qABCD characteristic is present in the imaged lesion, a report is provided to the display 20 d indicating that professional medical evaluation of this lesion should be obtained. Alternatively, the minimum number of qABCD characteristics can be increased to two or more. If no qABCD characteristics exist, the processor 20 c provides a report to the display 20 d indicating that the lesion is unlikely to be melanoma. The reports also may be uploaded through the network to a physician's computer.
In some implementations, the reports provided by the computer 20 also may include a display of the imaged lesion in color or in different spectral bands. In some implementations, the reports may include the image analysis data provided by the mathematical algorithms as well as the threshold values. In addition, the reports may compare the image analysis results for two or more lesions. For example, images of lesions which have been analyzed over a period of time may be displayed in a side by side comparison along with the corresponding image analysis data.
As explained above, mathematical algorithms are applied to the digital image to obtain the qABCD value for each qABCD characteristic. A discussion of examples of such algorithms may be found in U.S. Pat. No. 6,208,749. The lesion asymmetry (A) algorithm provides a measure of the geometric asymmetry of the lesion. To determine this value, a binary mask generated from the lesion image is used to evaluate the lesion orientation. Typically, the mask is generated during the image segmentation step (407) and stored in memory for later use.
If imaged lesions are to be compared over time or with other lesions, the lesion asymmetry value should be independent of the lesion orientation in the image. Therefore, the binary mask image is rotated by a lesion orientation angle θ to align the lesion principal axes (xp, yp) with the image axes (see
The border irregularity (B) algorithm is often defined as a ratio of a measured lesion perimeter to the perimeter of a circle having the same area as the lesion. However, because the perimeter is difficult to estimate reproducibly, a statistical descriptor of border irregularity may be used. In addition, many lesions are elongated and an ellipse may be a better approximation for the shape of the comparison lesion than a circle.
If a statistical descriptor is to be used for an elongated lesion, one first determines a lesion intensity centroid, an orientation angle, and an aspect ratio using the binary lesion mask. The aspect ratio is defined as
where the prime refers to the coordinate system defined by the lesion principal axes and the subscript c represents the lesion intensity centroid. In addition, the angular brackets denote the moment of intensity.
These values may then be used to construct an ellipse that is the best regular approximation of the lesion border. For each lesion border pixel, the angle between a line connecting this pixel to the lesion centroid and the horizontal axis is determined. A location on the ellipse border is determined for the same angle and a distribution of the distances between the ellipse border and the lesion border is computed. The border irregularity value may then be defined as the ratio of the standard deviation to the mean value of this distribution.
The color variegation (C) algorithm is defined as the ratio of the standard deviation of reflectance to the average reflectance over all of the pixels of the lesion in the spectral band centered at 700 nm. A wavelength of 700 nm is used because the pixel-to-pixel variability of lesion reflectance has been shown to have a maximum in the red spectral band, i.e., for wavelengths between 650 and 700 nm. However, since the spectral band of the red sensor or filter may not be centered at 700 nm, other wavelengths may be used as well.
The lesion dimension (D) algorithm is defined as the maximum distance (in millimeters) between two pixels on a lesion border instead of the lesion diameter because lesions are rarely circular.
Each of the qABCD values obtained using the above algorithms on an imaged pigmented lesion is evaluated against a corresponding threshold value to determine whether or not a qABCD characteristic is present. The threshold values may be pre-calculated for the specific imaging system and stored on the device 2 before first use. A pre-calculated threshold value is calculated as the qABCD value which maximizes a diagnostic accuracy of a corresponding qABCD characteristic in a sample database containing previously imaged and biopsied pigmented skin lesions. The sample database of biopsied lesions includes both clinically evaluated malignant melanomas and benign lesions. The diagnostic accuracy may be defined as DA=TP/(TP+FN+FP) where TP is the number of true positives (correctly identified melanoma in the sample database), FN is the number of false negatives (malignant lesions incorrectly identified as benign in the sample database) and FP is the number of false positives (benign lesions incorrectly identified as malignant in the database). However, different imaging devices may have different spatial resolutions, different illuminators that provide varying spectra of light, different sensor efficiencies, and different thermal noise. Therefore, the same threshold values may not be used for all imaging devices and may need to be determined independently for each imaging system. In addition, updated threshold values may be obtained by downloading them over the network connection from a server or other computer.
Clinical lesion characteristics other than qABCD can be evaluated by the device 2, as well. For example, an alternative set of lesion characteristics known as qRING characteristics may be evaluated in a similar manner to the qABCD characteristics. The RING acronym stands for the following parameters: mean lesion reflectance (R); lesion inhomogeneity measured by blotchiness (I); lesion network irregularity measured by irregularity of rete ridges (N), in which a rete ridge is an epidermal thickening that extends downward between dermal papillae; and gradient of the lesion border (G). The value for the mean lesion reflectance is determined using a digital image recorded in a red spectral band. The values for lesion inhomogeneity, network irregularity and border gradient are determined using digital images recorded in a blue spectral band. Digital images in both red and blue spectral bands may be obtained from the red and blue channels of the sensor in the camera 10. Similar to qABCD, quantitative RING (qRING) characteristics may be verified by comparing qRING values to corresponding threshold values, in which the threshold values are pre-calculated using a sample database. If one or more qRING characteristics are determined to be present in the lesion image, a report may be provided to the display 20 d indicating that professional medical evaluation should be obtained. Other lesion characteristics may analyzed as well.
Although the qRING characteristics do not correspond directly to visible characteristics that professionals have traditionally used to evaluate skin lesions, the qRING characteristics provide as good sensitivity and better specificity than qABCD for lesions that have been clinically evaluated.
Although the characteristics specified above are used to identify possible melanoma, the device 2 can also identify other pigmented and non-pigmented skin lesions that should be evaluated by a physician including, for example, actinic keratosis, psoriasis, basal cell carcinomas, and squamous cell carcinomas.
Certain implementations may have one or more of the following advantages. The lesion analysis device objectively and automatically determines lesion characteristics and makes straightforward comparisons to clinical evaluations of the characteristics. Furthermore, the lesion analysis device can be used as an objective tool for lesion identification. In addition, the device can provide a preliminary analysis of pigmented skin lesions without requiring evaluation by a dermatologist or other physician.
Other implementations and features are within the scope of the following claims.
For example, other conditions of interest of a person's skin may be imaged and the images analyzed quantitatively to determine whether additional analysis of an expert would be desirable. Such conditions include other health related conditions of the skin as well as other conditions of the skin that are not health related, for example, cosmetic conditions related to aging or beauty. Any condition on the surface or inside the skin that is of interest could be subject to such analysis. In each case, statistical data about the general population could be used as the basis to make a quantitative determination that is not necessarily accurate enough to be definitive but is accurate enough to suggest the need for an expert review. For example, a user could be advised whether a health-related or cosmetic skin condition would be susceptible to treatment by any chemical composition or whether further evaluation by an expert would be useful to make a final determination of that question.
Although in the examples given in the discussion above, the device is shown as a camera, an illuminator, and a computer, a wide variety of other implementations that use commercially available equipment may be possible now and in the future. It may be possible to do the processing directly in the digital camera and to display the result on the screen of the camera. The illuminator may be unnecessary in certain circumstances. The computer could be replaced by a variety of handheld devices. Eventually, it may be possible to use the digital camera and on-board processing capability of a cellular telephone to perform the entire process and display the result on the screen of the telephone.
In some implementations, the device may display an image of a body-map on which a user can identify a region that has been analyzed or a region that will be analyzed by the device. Moreover, a touch-sensitive display may allow a user to select the lesion location on the body-map image.
Although some of the examples discussed earlier refer to the use of the device either by a person in the general population or by a physician, including a dermatologist or a non-dermatologist, the user could have any of a wide variety of other roles, including, a nurse, a paramedical, a health care worker, a retail store clerk, a cosmetologist, a caregiver, a trainer, a coach, a pharmacist, a family member, or a friend, to name a few.
In addition, the knowledgeable person could be other than a dermatologist, for example, a non-dermatologist physician, a health care worker, or a cosmetologist, to name a few.
The person whose skin is being analyzed need not be a patient, as suggested in some of the implementations described earlier, but could be any person having any role in the general population, including an athlete, a performer, a teenager, a consumer, or a traveler, to name a few.
The contexts in which the analysis could be performed need not be limited to the locations previously named, but could include any place where it is convenient or desirable to do so, including, for example, an examination room, an outpatient clinic, a pharmacist's facility, a waiting room, an airport, a retail store, a school, a nurse's office, a place of employment, a cruise ship, a beach cabana, or a resort facility, to name a few.
The analysis device can be implemented in a wide variety of form factors, including hand held devices, desktop devices, devices installed in buildings, rooms, vehicles, and any other place where such an analysis would be desirable or convenient.
The specificity and sensitivity levels of the analysis can vary over a range and still provide useful information to the user.
A variety of quantitative analyses can be used to obtain the information to be provided to the user.
The population to which the classification applies need not be the general population, but could be for example, a population limited by various demographic characteristics, such as ethnic background, skin color, geography, sex, or age.
The parameters used in the analysis can all be ones that correspond to visible familiar skin characteristics, can all be ones that do not correspond to any familiar visible skin characteristics, or can be a mixture of the two.
A wide variety of display and audio techniques can be used to indicate to the user the result of the quantitative analysis.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US6208749 *||Feb 27, 1998||Mar 27, 2001||Electro-Optical Sciences, Inc.||Systems and methods for the multispectral imaging and characterization of skin tissue|
|US6215893||Aug 20, 1998||Apr 10, 2001||Romedix Ltd.||Apparatus and method for measurement and temporal comparison of skin surface images|
|US6648820||Oct 27, 1999||Nov 18, 2003||Home-Medicine (Usa), Inc.||Medical condition sensing system|
|US6770029 *||Feb 13, 2003||Aug 3, 2004||First Opinion Corporation||Disease management system and method including correlation assessment|
|US6792137||Feb 14, 2001||Sep 14, 2004||Robert Kenet||Method and device for skin cancer screening|
|US7162063 *||Jul 26, 2004||Jan 9, 2007||Western Research Company, Inc.||Digital skin lesion imaging system and method|
|US7647285 *||Nov 2, 2006||Jan 12, 2010||Microsoft Corporation||Tools for health and wellness|
|US20050171439||Jan 18, 2005||Aug 4, 2005||Michael Maschke||Method and device for examining the skin|
|US20050251415 *||May 7, 2004||Nov 10, 2005||Pak Hon S||System and method for consultation on dermatological disorders|
|US20060269111 *||May 30, 2006||Nov 30, 2006||Stoecker & Associates, A Subsidiary Of The Dermatology Center, Llc||Automatic detection of critical dermoscopy features for malignant melanoma diagnosis|
|US20080049990 *||Apr 15, 2004||Feb 28, 2008||Carmelo Francesco Melchi||Automatic Detection of Skin Lesions|
|US20080214907 *||Mar 2, 2007||Sep 4, 2008||Dina Gutkowicz-Krusin||Quantitative Analysis Of Skin Characteristics|
|1||"Poster Abstracts," Supplement to the Journal of the American Academy of Dermatology, 54(3) (Mar. 2006), American Academy of Dermatology 64th Annual Meeting, Mar. 3-7, 2006.|
|2||Abbasi, N. R., et al. "Early diagnosis of cutaneous melanoma. Revisiting the ABCD criteria." JAMA; 292:2771-2776 (2004).|
|3||Bono, A., et al. "The ABCD system of melanoma detection." Cancer, 85:72-77 (1999).|
|4||Dermlite: http://www.dermlite.com/fotosyspro.html , DemiLite FOTO + DermLite FOTOsystem, pp. 1-6, Mar. 6, 2007.|
|5||DermoGenius Basic II brochure, pp. 1-5.|
|6||DermoGenius http://www.dermogenious.com/DermoGenius-lite.70.0.html?&L=1, DermoGenius Lite, pp. 1-2, Mar. 2, 2007.|
|7||DermoGenius http://www.dermogenius.com/DermoGenius-MoldMap.78.0.html?&L=1, DermoGenius + MoleMap, p. 1, Mar. 2, 2007.|
|8||DermoGenius http://www.dermogenius.com/DermoGenius-MoleMap.78.0.html?&L=1, DermoGenious + MoleMap, p. 1, Mar. 2, 2007.|
|9||DermoGenius http://www.dermogenius.com/index.php?id=72&L=1, DermoGenius Ultra, pp. 1-2, Mar. 2, 2007.|
|10||DermoGenius http://wwwdermogenius.com/DermoGenius-basic-II.68.0html&L=1, DermoGenius Basic II, p. 1, Mar. 2, 2007.|
|11||Elbaum, M., et al. "Automatic differentiation of melanoma from melanocytic nevi with multispectral digital dermoscopy: a feasibility study." J Am Acad Dermatol, 44:207-218 (2001).|
|12||Friedman, R. J., et al. "Early detection of malignant melanoma: the role of physician examination and self-examination of the skin." CA Cancer J Clin, 35:130-151 (1985).|
|13||Gutkowicz-Krusin, D., et al. "Evaluation of clinical and quantitative ABCD characteristics of pigmented skin lesions in the diagnosis of melanoma." Presented at the American Academy of Dermatology Annual Meeting, San Francisco, CA. Mar. 3, 2006 (abstract).|
|14||Gutkowicz-Krusin, D., et al. "Precision of automatic measurement of parameters of pigmented skin lesions with a multispectral digital dermoscope MelaFind." Melanoma Res, 10:563-570 (2000).|
|15||International Search Report, Patent Cooperation Treaty, dated Jun. 25, 2008 (10 pages).|
|16||Kopf, A. W., et al. "Atypical mole syndrome." JAAD, 22:117-118 (1990).|
|17||Schaefer, T., et al. "The epidemiology of nevi and signs of skin aging in the adult general population." J Invest Dermatol, 126:1490-1496 (2006).|
|18||Siascope: http://vvww.astronclinica.com/products/moleview.htm, MoleView, Mar. 2, 2007, pp. 1-4.|
|19||Siascope: http://www.astronclinica.com/products/basal-cell-carcinoma.htm, BCC View, Mar. 2, 2007, pp. 1-3.|
|20||Siascope: http://www.astronclinica.com/products/contact-siascopy.htm, SlAscope V-Contact SlAscopy, Mar. 2, 2007, pp. 1-3.|
|21||Siascope: http://www.astronclinica.com/products/contact-siascopy.htm, SlAscope V—Contact SlAscopy, Mar. 2, 2007, pp. 1-3.|
|22||Siascope: http://www.astronclinica.com/products/molemanager.htm, MoleManager, Mar. 2, 2007, pp. 1-3.|
|23||Siascope: http://www.astronclinica.com/products/molemate.htm, MoleMate, Mar, 2, 2007, pp. 1-5.|
|24||Siascope: http://www.astronclinica.com/products/moleview.htm, MoleView, Mar. 2, 2007, pp. 1-4.|
|25||Siascope: http://www.astronclinica.com/products/non-contact-siascopy.htm, Digital Camera-Non-Contact SIAscopy, Mar. 2, 2007, pp. 1-2.|
|26||Siascope: http://www.astronclinica.com/products/non-contact-siascopy.htm, Digital Camera—Non-Contact SIAscopy, Mar. 2, 2007, pp. 1-2.|
|27||Siascope: http://www.astronclinica.com/technology/siascopy-explained.htm, What is SIAscopy?, Mar. 2, 2007, pp. 1-4.|
|28||SolarScan: http://www.polartechnics.com.au/proSolar.htm Mar. 2, 2007.|
|29||SolarScan: http://www.polartechnics.com.au/SolarTech.htm, pp. 1-3, Mar. 2, 2007.|
|30||Thomas, L., et al. "Semiological value of ABCDE criteria in the diagnosis of cutaneous pigmented tumors." Dermatology, 197:11-17 (1998).|
|31||Tsao, H., et al. "The transformation rate of moles (melanocytic nevi) into cutaneous melanoma." Arch Dermatol, 139:282-288 (2003).|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US8170925||Jul 18, 2011||May 1, 2012||Nor1, Inc.||Computer-implemented methods and systems for automatic merchandising|
|US8285599||Mar 28, 2012||Oct 9, 2012||Nor1, Inc.||Method and a system for simultaneous pricing and merchandising|
|US9075906||Jun 28, 2013||Jul 7, 2015||Elwha Llc||Medical support system including medical equipment case|
|US9089303 *||Nov 12, 2012||Jul 28, 2015||Lubax, Inc.||Automated assessment of skin lesions using image library|
|US9101320||Apr 9, 2013||Aug 11, 2015||Elc Management Llc||Skin diagnostic and image processing methods|
|US20130218026 *||Nov 12, 2012||Aug 22, 2013||Convergent Medical Solutions, Inc.||Automated assessment of skin lesions using image library|
|USD743553||Feb 28, 2013||Nov 17, 2015||DermSpectra LLC||Imaging booth|
|U.S. Classification||382/128, 606/9|
|Cooperative Classification||A61B5/444, A61B5/445, A61B2560/0233, A61B5/0059|
|European Classification||A61B5/00P, A61B5/44B4, A61B5/44B6|
|May 2, 2007||AS||Assignment|
Owner name: ELECTRO-OPTICAL SCIENCES, INC., NEW YORK
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:GUTKOWICZ-KRUSIN, DINA;GULFO, JOSEPH V.;REEL/FRAME:019237/0441;SIGNING DATES FROM 20070417 TO 20070426
Owner name: ELECTRO-OPTICAL SCIENCES, INC., NEW YORK
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:GUTKOWICZ-KRUSIN, DINA;GULFO, JOSEPH V.;SIGNING DATES FROM 20070417 TO 20070426;REEL/FRAME:019237/0441
|May 7, 2010||AS||Assignment|
Owner name: MELA SCIENCES, INC.,NEW YORK
Free format text: CHANGE OF NAME;ASSIGNOR:ELECTRO-OPTICAL SCIENCES, INC.;REEL/FRAME:024351/0849
Effective date: 20100430
Owner name: MELA SCIENCES, INC., NEW YORK
Free format text: CHANGE OF NAME;ASSIGNOR:ELECTRO-OPTICAL SCIENCES, INC.;REEL/FRAME:024351/0849
Effective date: 20100430
|Aug 18, 2014||FPAY||Fee payment|
Year of fee payment: 4